History

30 days of forecast vs. what actually happened at the lake. Click a single day to see its details.

Last 30 days

Rule-based forecast (actually used)
ML-based forecast โ€” logistic (experimental)
ML-based forecast โ€” HistGradientBoosting (black-box model, experimental)
Actual measurements (session โ‰ฅ 1 h)
Thu Jun 25 Fri Jul 24
session (โ‰ฅ 1 h โ‰ฅ 11 kt) marginal (โ‰ฅ 1 h โ‰ฅ 8 kt) no wind no data thunderstorm (forecast rows: predicted ยท actual row: observed)

Jul 1, 2026

81106:0021:0007:26 ยท avg 2.0 kt ยท gust 4.9 kt07:58 ยท avg 2.5 kt ยท gust 4.3 kt08:27 ยท avg 0.7 kt ยท gust 2.2 kt09:29 ยท avg 4.7 kt ยท gust 6.5 kt09:32 ยท avg 5.8 kt ยท gust 7.6 kt09:59 ยท avg 11.1 kt ยท gust 16.2 kt10:04 ยท avg 8.9 kt ยท gust 17.3 kt11:09 ยท avg 5.0 kt ยท gust 8.6 kt11:10 ยท avg 7.6 kt ยท gust 8.6 kt14:29 ยท avg 8.8 kt ยท gust 10.8 kt14:34 ยท avg 9.1 kt ยท gust 10.8 kt15:18 ยท avg 12.2 kt ยท gust 14.6 kt15:49 ยท avg 6.8 kt ยท gust 10.8 kt18:19 ยท avg 5.0 kt ยท gust 10.3 kt18:21 ยท avg 8.6 kt ยท gust 13.0 kt19:03 ยท avg 10.3 kt ยท gust 11.9 kt19:19 ยท avg 8.7 kt ยท gust 13.5 kt19:26 ยท avg 7.7 kt ยท gust 11.9 kt
avg wind gust ignition (8 kt) session (11 kt)
For Jul 1, 2026 ยท as written at the time

NO GO

100% daytime cloud + only 92 W/mยฒ solar โ€” no insolation to drive a thermal

โšก Storm risk: the thermal often runs well until the front โ€” watch the sky, come in early.

๐Ÿค– ML โ€” logistic experimental NO GO GO 10% ยท MAYBE 14% ยท NO GO 76%

Strongest inputs โ€” (+) dew-point spread; (โˆ’) morning solar, pressure ฮ” (thermik).

๐Ÿค– ML โ€” HGB ยท black box experimental MAYBE GO 0% ยท MAYBE 53% ยท NO GO 47%

Most distinctive inputs: morning solar, daytime cloud, Innsbruck pressure.

Experimental shadow models โ€” they never drive the verdict. How good are they? โ†’

Previous forecast method

How the day was forecast at the time, before the rules were recalibrated โ€” shown for comparison only.